BON-10553; No. of pages: 9; 4C:
Bone xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Bone
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b o n e
1
2
3
Original Full Length Article
MicroRNA-20a regulates autophagy related protein-ATG16L1 in
hypoxia-induced osteoclast differentiation
Q54Q4 Kuo-Ting
5
6
7
8
9
10
11
12
a
Sun a,b,c, Michael Y.C. Chen c,d, Ming-Gene Tu c, I-Kuan Wang a,e, Shih-Sheng Chang a,f, Chi-Yuan Li a,g,⁎
Graduate Institute of Clinical Medical Science, China Medical University, No. 91 Hsueh-Shih Rd., Taichung, Taiwan
b
Department of Pediatric Dentistry, China Medical University Hospital, No. 2 Yu-Der Rd., Taichung, Taiwan
c
School of Dentistry, China Medical University, No. 91 Hsueh-Shih Rd., Taichung, Taiwan
d
Department of Oral & Maxillofacial Surgeon, China Medical University Hospital, No. 2 Yu-Der Rd., Taichung, Taiwan
e
Division of Nephrology, Department of medicine, China Medical University Hospital, No. 2 Yu-Der Rd., Taichung, Taiwan
f
Division of Cardiology, Department of medicine, China Medical University Hospital, No. 2 Yu-Der Rd., Taichung, Taiwan
g
Department of Anesthesiology, China Medical University Hospital, No. 2 Yu-Der Rd., Taichung, Taiwan
article
info
abstract
13
Article history: Autophagy and autophagy-related proteins (ATGs) play decisive roles in osteoclast differentiation. Emerging 27
14
Received 8 October 2014
15
Revised 25 November 2014
19
Edited by Hong-Hee Kim
lines of evidence show the deregulation of miRNA in autophagic responses. However, the role of hypoxia and in- Q728
volvement of miRNA in osteoclast differentiation are unclear. In the present study, we demonstrate that hypoxia 29 16 Accepted 29 November 2014 caused
induction of autophagy and osteoclast differentiation markers in RAW264.7 cells stimulated with M-CSF 30
17 Available online xxxx and RANKL. In addition, miR-20a was significantly repressed during hypoxia and identified as the prime candi- 31
18
UNCORRECTED PROOF
date involved in hypoxia-induced osteoclast differentiation. The results from dual luciferase reporter assay re- 32 vealed that miR-20a directly targets
Atg16l1 by binding to its 3′UTR end. Further, miR-20a transfection studies 33
20
Keywords:
showed significant down regulation of autophagic proteins (LC3-II and ATG16L1) and osteoclast differentiation 34
21
Osteoclast
markers (Nfatc1, Traf6, and Trap) thus confirming the functional role of miR-20a under hypoxic conditions. Re- 35
22
Autophagy
sults of chromatin immunoprecipitation assay showed that HIF-1α binds to miRNA-20a. From miRNA Q-PCR re- 36
Hypoxia
sults, we confirmed that shRNA HIF-1α knockdown significantly downregulated both autophagy (LC3, p62, Atg5, 37
23
24
HIF-1α
Atg12, Atg16l1, Atg7, Becn1, Atg9a) and osteoclast markers (Traf6, Nfatc1, Ctsk, cFos, Mmp9, Trap) in RAW264.7 38 microRNAs findings suggest that the regulatory axis of HIF-
1α-miRNA-20a-Atg16l1 might be a critical mech- 25
39
Periodontal disease
40
41
4523
44
cells. Thus, our
26
anism for hypoxia-induced osteoclast
differentiation.
© 2014 Published by Elsevier Inc.
46 1. Introduction autophagosome formation [3]. Studies confirm that deletion of autoph- 61 agic proteins such as ATG5 and ATG7 was found to impair the ruffled 62 47 Bone
homeostasis is a dynamic process involving a delicate balance border of osteoclasts and reduce osteoclastic bone resorption [4]. Au- 63
between formation and resorption. The osteoclasts and osteoblasts are tophagy and ATGs play a critical role in osteoclast differentiation. 64
49
the major cells involved in bone remodeling [1]. However, when this
Thus, regulation of osteoclast differentiation can be achieved by 65
50
balance shifts towards increased osteoclast activity, it ultimately results targeting autophagy. So, it is important to understand these intriguing 66
51
in bone loss leading to bone-related inflammatory diseases such as peri- mechanisms to develop potential strategies against bone-related in- 67
52
odontal disease, rheumatoid arthritis and osteoporosis [2]. Bone- flammatory diseases. 68
53
resorbing osteoclasts are multinuclear giant cells, which are derived
MiRNAs are a novel class of small, single-stranded non-coding RNAs 69
54
from hematopoietic cells of the monocyte/macrophage lineage in the
that post-transcriptionally regulate gene expression via mRNA degrada- 70
55
presence of RANKL (receptor activator of NF-kB ligand) and M-CSF tion or translational inhibition of protein synthesis. Recent studies sug- 71
56
(macrophage colony stimulating factor) [1]. Initiation and progression gested that miRNAs participated in osteoclast differentiation [5]. MiR- 72
57
of osteoclast differentiation of hematopoietic cells are manipulated
223 is highly expressed in rheumatoid arthritis synovium, and over ex- 73
58
by genetic and epigenetic mechanisms. Autophagy is the principal
pression of miR-223 could suppress osteoclastogenesis [6]. MiR-503 74
59
catabolism for degradation of dysfunctional cellular components. was shown to inhibit receptor activator of RANKL-induced osteoclasto- 75
48
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
2
60
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
The autophagy-related proteins (ATGs) are known to engage in genesis in CD14(+) peripheral blood mononuclear cells [7]. MiR-124 76 regulates proliferation and
motility of osteoclast precursor macro- 77
phages by suppressing nuclear factor of activated T-cells, cytoplasmic, 78
⁎ Corresponding author at: Graduate Institute of Clinical Medical Science, China Medical
Q6
calcineurin-dependent 1 (NFATc1) expression [8]. MiR-21, -29b, -146a 79
and -210 all were proved to participate in the differentiation of 80
University, No. 91 Hsueh-Shih Rd., Taichung, Taiwan. Fax: +886 4 22333710.
E-mail address: cyli168@gmail.com (C.-Y. Li).
http://dx.doi.org/10.1016/j.bone.2014.11.026 8756-3282/©
2014 Published by Elsevier Inc.
81 RAW264.7 cells to osteoclast-like cells after treatment with TNF-α/ for 6 days. Cells (1 × 105) were seeded onto 10-cm2 dishes for western 139 82 RANKL [9]. Taken
together, miRNAs are involved in regulation of osteo- blot, real-time PCR analysis, reporter gene assay and mimic transfection. 140 83 clast differentiation.
84 Hypoxia acts as a stimulator of osteoclast formation and bone re- 2.4. TRAP (Tartrate Resistant Acid Phosphatase) staining 141 85 sorption [10,11]. Autophagy and ATGs
may mediate osteoclast differen86
tiation induced by hypoxia through hypoxia-inducible factor 1-alpha
Important characteristic feature and markers for the identification of 142
87
(HIF-1α)/BCL2/adenovirus E1B 19kd-interacting protein 3 (BNIP3)
OC differentiation include TRAP expression and the presence of multi- 143
88
pathway [12]. More recent research shows miRNA as key regulators in
nucleus [12]. Briefly, after respective treatment schedule, the cells 144
89
autophagy [13,14]. However, the role of miRNAs on autophagy during
were treated with TrypLE™ and fixed with 3.8% paraformaldehyde for 145
90
hypoxia and its subsequent impact on hypoxia-induced osteoclast
10 min followed by ethanol/acetone (1:1) fixation for 1 min. Post- 146
91
differentiation has not been explored. The present study was de-
fixation, the cells were stained with TRAP staining solution for 1 h at 147
92
signed to investigate whether and how miRNAs serve an important 37 °C and observed under a NIKONT100 inverted contrasting micro- 148 93 role in autophagy
mediating osteoclast differentiation under hypox- scope. TRAP positive cells with more than 3 nuclei were counted as ma- 149
94
95
96
97
98
99
ia circumstances. ture osteoclasts. 5 random fields were selected and counted for mature 150 osteoclasts and their average is represented as OCs/field.
2. Materials and methods
2.5. Prediction of miRNA target genes
2.1. Reagents
151
152
The miRWalk database was used to predict the binding site on the 3′ 153
Recombinant human soluble RANKL and M-CSF were purchased
UTR of Atg16l1 gene; this database combines several bioinformatic plat- 154
from PeproTech (Rocky Hill, NJ). Dharmacon miR-20a miRNAmimics,
forms including TargetScan 4.2 (http://targetscan.org), miRBase, 155
anti-miR-20a miRNA inhibitors (AntagomiR) and Scramble were pur- RNA22, DIANA-mT, miRDB, PICTAR5, PITA, and miRanda.
156 100 chased
from
Dharmacon (Denver, CO). Roche UPL microRNA assay for
101 mmu-mir-20a, miRNA reverse transcription kit and universal PCR mas- 2.6. Construction of miRNAs and reporter vectors 157 102 ter mix was purchased from Roche
(Welwyn Garden City, UK).
TrypLE™, Lipofectamine 2000 transfection reagents were from Life
miRNAs and reporter gene construction was performed as described 158
104
Technologies (Carlsbad, CA). ChIP assay kit was purchased from
previously [15,16]. Genomic fragments of miR-20a precursors were am- 159
105
Millipore (Darmstadt, Germany). IgG or anti-HIF-1α antibody from
plified by PCR using mouse genomic DNA as a template (the PCR 160
106
Novus Biologicals (Southpark Way, USA). pmirGLO luciferase vector primers are listed as Supplementary data—Table 1). The PCR products 161
107
from Promega (Madison, WI). Trizol Reagent, random hexamer primers,
were cloned into the pLAS2-RFP vector at restriction sites NotI and 162
108
First Strand Buffer, RNaseOUT, dithiothreitol, MMLV Reverse Transcrip-
XhoI. RAW264.7 cells were virally infected with the miRNAs, the ex- 163
109
tase, FITC and HRP-conjugated goat anti-rabbit IgG from Invitrogen
110
(Carlsbad, CA).
111
ATG16L1 (NB110-60928), p62 (NBP1-48320SS) and ATG5 (NBP110- target genes and the entire Atg16l1 3′UTR sequence was cloned into 166
112
87318) were procured from Novus Biologicals (Southpark Way, USA). the pmirGLO luciferase vector at restriction sites PmeI and XhoI. Muta- 167
113
Anti β-actin (AC-74) antibody was from Sigma (St. Louis, MO). All
114
other chemicals were purchased from Sigma (St. Louis, MO). tant constructs of the Atg16l1 3′UTR were generated by a pair of 169 primers containing the mutant
Primary
antibodies
against LC3
NCORRECTED PROOF
103
pressions of which were then detected with the aid of quantitative 164
(NB100-2220),
PCR (Q-PCR) (ABI Step-One Plus). The binding site for miR-20a in the 165
tion constructs were performed as described previously [17]. The mu- 168
sequence. 170
115
2.2. Cell culture and treatments
116
RAW264.7 mouse macrophage cells were purchased from American
117
Type Culture Collection (ATCC-TIB71). The cells were grown in
118
Dulbecco's modified eagle's medium (Gibco, Invitrogen) supplemented agent (Invitrogen) according to the manufacturer's instructions, and 173
119
with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. Cells
120
were used at 5 to 10 passages. Cells were seeded 24 h before the treat- trophotometer (Thermo Scientific Waltham, MA,). For reverse tran- 175
121
ment. Cells in α-minimal essential medium supplemented with 10% FBS
122
123
and 1% penicillin/streptomycin were exposed to normoxia (20% O2) or 10 ml of ultrapure water (free of DNase and RNase); 1 μl of 50 mM ran- 177
hypoxia (0.3% O2) in the presence of 100 ng/ml RANKL and 20 ng/ml dom hexamer primers was added. The solution was denatured at 65 °C 178
124
M-CSF for 24 h and were used for determination of autophagy and oste-
2.7. Reverse transcription of RNA and Q-PCR
Total RNA was extracted from the RAW264.7 cells with Trizol Re- 172
the RNA concentration was determined using a NanoDrop 2000 spec- 174
scription of RNA to first-strand cDNA, 1 μg of RNA was diluted in 176
for 5 min and 4 °C for 1 min, followed by the addition of 4 μl of 5× First 179
Strand Buffer, 1 μl of 10 mM dNTP, 1 μl of RNaseOUT, 2 μl of 0.1 M dithio- 180
were transfected with control or miR-20a mimic or antagomiR for threitol, and 1 ml of MMLV Reverse Transcriptase. The mixture was in- 181
125Q8 oclast differentiation markers. For miR-20a mimic hypoxia study, cells
126
171
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
3
127
24 h. Followed by which the cells were trypsinized, plated and allowed
cubated at different temperatures and time points sequentially (25 °C, 182
128
to settle for 4 h. Cells were then subjected to hypoxic treatment for 24 h
and after refreshing, osteoclast differentiation medium was added and
10 min; 45 °C, 60 min; 75 °C, 5 min).
exposed to normoxic conditions for 6 days. After the complete treatment schedule, cells were lysed and used for further analysis.
2.8. Western blot analysis
129
130
131
132
134
183
184
Cell lysis was performed using lysis buffer (Thermo Scientific, 185
2.3. Osteoclast differentiation Waltham, MA). Samples with equal protein concentration were subject- 186 ed to SDS-PAGE and transferred to a PVDF membrane
(PerkinElmer, Life 187 133 Cells were seeded 24 h before treatment. RAW264.7 cells were Science). Blots were blocked with 5% skim milk–TBS–Tween 20 for 1 h 188
grown in α-minimal essential medium supplemented with 10% FBS,
at room temperature and probed with primary antibodies (1:500) 189
1% penicillin/streptomycin for 7 days under normoxic (20% O2) condiagainst LC3, ATG16L1, ATG5, p62 and β-actin overnight at 4 °C. Followed 190
tion in the presence of 100 ng/ml RANKL and 20 ng/ml M-CSF. For hyp- by which blots were washed with PBS–Tween 20 and incubated with 191
137
oxia, the cells were exposed to hypoxia for 24 h, followed by which horseradish peroxidase-conjugated secondary antibodies (1:1000) for 192
138
medium was refreshed and incubated in normoxic (20% O2) conditions
1 h at room temperature. The blots were washed and incubated in ECL 193
194
244
solution (Thermo Scientific, Waltham, MA) for 1 min and then exposed by 2.15. Statistical analysis
195
ImageQuant LAS4000 (GE Healthcare).
All the experiments were performed as 3 independent experiments 245 in
triplicates. The mean and standard deviation were calculated for each 246 of the
196
2.9. Transfection of miRNA mimics
determined parameters. Error bars represent standard deviation 247 (SD) of a
triplicate set of experiments. Statistical analyses were per- 248 formed using
197
The miRIDIAN miRNA mimics are single-stranded chemically enhanced
unpaired Student's t test and ANOVA. The level of statisti- 249 cal significance was
198
oligonucleotides that were designed to mimic miRNA overexpression.
set at P b 0.05 and P b 0.01. 250
RAW264.7 cells were transfected with 100 nM of either the miR-20a mimics
199
135
136
200
201
or scramble mimics using the Lipofectamine 2000 reagent after 24 h, cells
were plated for the luciferase reporter assay.
202
2.10. miRNA Q-PCR for target gene validation
3.1.
Hypoxia
203
Real-time PCR analysis was performed as described elsewhere [18].
Real-time PCR was conducted using universal reverse primer, miRNA-
3. Results
induces
autophagy
251
and
ATG16L1
during osteoclast 252
differentiation
205
206
208
209
210
211
212
213
214
To determine whether autophagy could be induced by hypoxia during 254
specific forward primers, 2× Master mix and UPL probe-21. osteoclast differentiation, RAW264.7 cells were treated with 100 ng/ml 255
RANKL and 20 ng/ml M-CSF under hypoxia and normoxia conditions. 256
After 7 days of treatment (1 day of hypoxia with subsequent treatment 257
2.11. shRNA knockdown by viral infection under normoxia), a marked increase in the number of multinucleated 258 osteoclast-like cells was observed through
TRAP staining (Fig. 1A). Fur- 259 207 Human embryonic kidney 293T (HEK293T) cells were cultured as thermore, immunofluorescence staining for LC3 (a
major marker of au- 260
described elsewhere [19]. RAW264.7 cells were infected with lentivirus tophagy) showed increase in LC3 expression under hypoxia but not 261
expressing shRNA for HIF-1α (TRCN0000232221) in the presence of
8 μg/ml protamine sulfate for 24 h, followed by puromycin (2 μg/ml;
48 h) selection. shLacZ (TRC0000231726), which targets the LacZ
under normoxia (Fig. 1B). The expression levels of Traf6, an early mark- 262
er of osteoclast differentiation increased under hypoxia (Fig. 1C). In ad- 263
dition, the levels of Nfatc1, Ctsk, cFos and Mmp9 were upregulated in 264
CORRECTED PROOF
204
253
gene, was used as a control. The knockdown efficiency of HIF-1α was hypoxia compared to normoxia (Fig. 1C). The mRNA expression of 265
examined using RT-PCR (Biometra T1 thermo cycler). autophagy-related proteins such as LC3, Atg5, Atg12, Atg7 and Atg16l1 266 was upregulated under hypoxic
conditions (Fig. 1D). Increased levels 267 of LC3 and ATG16L1 were further confirmed by western blotting 268
2.12. Immunostaining (Fig. 1E). These findings suggest that hypoxia could induce osteoclast 269 differentiation and expression of autophagy-related proteins,
including 270 215 After the treatment schedule, RAW264.7 cells were fixed in 4% form- ATG16L1 during osteoclast differentiation (Fig. 1). 271 216 aldehyde.
Followed by which cells were then washed with PBS, blocked 217 with 1% BSA and stained with primary antibodies against LC3 or
218 ATG16L1 (1:100) for 60 min and incubated with FITC-conjugated sec- 3.2. Hypoxia reduces miR-20a expression and miR-20a modulates Atg16l1 272 219 ondary
antibodies. The nuclei were counter stained with DAPI for
220
30 min. Immunofluorescence of cells was visualized under a Zeiss fluo- Atg16l1 gene is an important player in autophagosome formation. To 273
221
rescence microscope with excitation (ex.) and emission (em.) wave-
222
length, 488 nm and LP 505 respectively. Images captured were TargetScan 4.2 (http://targetscan.org) was used [20]. From the results 275 223 corrected for background
investigate miRNA target of Atg16l1 gene, bioinformatics platform— 274
and analyzed for fluorescence intensity seven miRNA candidates were selected: miR-19a-3p, -20a-5p, -93a- 276
224
using Zen software (Ziess). 5p, -96-5p, -106b-5p, -130a-3p, and -142-3p. Next, we determined 277 the expression pattern of the 7 candidate miRNAs in
RAW264.7 cells 278
225
2.13. Dual-luciferase reporter assay at an early stage (24 h) of commitment to osteoclast-like cells. The ex- 279 pression patterns of candidate miRNAs that may
suppress osteoclast dif- 280
226
RAW264.7 cells at 50% confluence were seeded in six-well plates
227
and
allowed
normoxia con- 282
228
transfection was performed with miR-20a mimics and Atg16l1 3′UTR re-
dition (Fig. 2A). The level of miR-20a in osteoclast differentiation de- 283
229
porter vector using Lipo2000; 1 μg of Atg16l1 3′UTR or control vector
creased significantly at days 1, 3 and 5 after 24 h hypoxia treatment 284
for
24
h
ferentiation; we focused on miR-20a, which was down-regulated under 281
differentiation. Followed
by
which, co-
hypoxia-induced
osteoclast
differentiation
compared
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
to
4
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
230
and 100 nM miR-20a mimics or scramble mimics per well. Cell extracts
231
were prepared at 24 h after transfection and luciferase activity was
232
measured using the Dual-Luciferase Reporter Assay System (Promega,
233
Madison, WI). 20a during osteoclast differentiation. 288
234
235
236
237
238
239
240
241
242
243
(Fig. 2B). In addition, ATG16L1 was up-regulated during hypoxia- 285
induced osteoclast differentiation as observed through immunofluores- 286
2.14. Chromatin immunoprecipitation (ChIP) assay
ChIP was performed according to the manufacturer's instructions (Magna
ChIP A/G Chromatin Immunoprecipitation Kit, Millipore, Darmstadt,
Germany). Briefly, after the treatment schedule, RAW264.7 cells were fixed
with 1% paraformaldehyde and lysed by SDS lysis buffer. The cell lysate was
sonicated and protein-DNA complexes were immuno-precipitated with
control IgG or anti-HIF-1α antibody. Followed by this, protein/DNA
complexes were eluted, reverse cross-linked to free DNA and purified. Q-PCR
was performed with primers specific for mouse miR-20a promoter.
cence (Fig. 2C). All these data suggested that hypoxia could repress miR- 287
According to the bio-informatics database, we predicted that miR- 289
20a binding site was at the 3′-untranslated region (UTR) of Atg16l1 290
(Fig. 2D). To validate whether miR-20a directly binds to 3′UTR of 291 Atg16l1,
we constructed a full length of 3′UTR of Atg16l1, which have 292 highly
conserved binding site by miR-20a, and subcloned into a pmir- 293 GLO dual
luciferase reporter at the 3′ end of firefly luciferase coding se- 294 quence as
3′UTR of Atg16l1 wild-type (WT), and mutated binding se- 295 quence from
GCACTTTA to TTGACCC as 3′UTR of Atg16l1 mutant (MT) 296 (Fig. 2E).
Luciferase reporter assay showed that transfection of miR- 297
20a mimics significantly repressed the activity in the 3′UTR of 298 Atg16l1-WT
compared with 3′UTR of Atg16l1-MT (Fig. 2E). These results 299 confirmed the
specificity of the miR-20a modulation to the 3′UTR of 300
Atg16l1.
301
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
UNCORRECTED PROOF
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
5
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
Fig. 1. Hypoxia
autophagy during
differentiation in
cells.
(A).
cells cultured in
presence
of
and
M-CSF
0.3% hypoxia for
followed
by
for
6
days)
increase
in
positive
multinucleated
are mean ± SD,
representative of
independent
experiments
in triplicates. *P b
compared
to
Magnification
scale bar = 100
Immunostaining
Hypoxic
(24 h) enhanced
levels.
LC3
visualized
as
whereas
the
stained
with
appears in blue.
compared
to
Magnification
scale bar = 20 μm.
representative
obtained
from
individual
experiments. qRT(C)
showing
mRNA expression
osteoclast
differentiation
after 24 h hypoxia.
Expression
of
genes after 24 h
The data are
the mean ± SD of
independent
experiments
in triplicates. *P b
compared
to
(E).
Hypoxic
increases
Q1 autophagic response
of proteins LC3-I,
ATG5 and p62 in
manner
in
The
relative
expressions were
β-actin.
(For
the references to
legend, the reader
web version of
Fig. 2. Atg16l1
miR-20a
differentiation
miRNA
Qcandidate
RAW264.7
hypoxia; (B).
Q2 at differentiation
3 and 5) post 24
data are shown
SD of three
experiments
triplicates. *P b
to normoxia. (C).
Immunostaining:
cells
were
ATG16L1 levels
osteoclast
under hypoxia
ATG16L1
visualized
as
whereas
the
with
DAPI
A representative
from
three
experiments.
(100×);
scale
increases
osteoclast
RAW264.7
RAW264.7
the
RANKL
(under
24
h
normoxia
showed
TRAPcells. Data
three
performed
0.05
normoxia.
(100×);
μm.
(B).
for
LC3:
treatment
the
LC3
protein
green color;
nucleus
DAPI
*P b 0.05
normoxia.
(200×);
A
image
three
PCR results
enhanced
of
UNCORRECTED PROOF
6
markers
(D).
autophagy
hypoxia.
shown as
three
performed
0.05
normoxia.
treatment
through regulation
LC3-II, ATG16L1,
a time dependent
RAW264.7 cells.
protein
determined against
interpretation of
color in this figure
is referred to the
this article.)
is a direct target for
during
osteoclast
under
hypoxia.
PCR:
(A).
7
miRNAs
of
cells post 24 h
miRNA-20a
time (days 0, 1,
h hypoxia. The
as the mean ±
independent
performed in
0.05 compared
ATG16L1
RAW264.7
monitored for
at day 7 during
differentiation
(24
h).
protein
green
color,
nucleus stained
appears in blue.
image obtained
individual
Magnification
bar = 100 μm.
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
7
*P b 0.05 compared to normoxia. (D). Base pairing comparison between mature miR-20a and wild type or mutant Atg16l1 3′UTR putative target site is shown according to the TargetScan database. (E).
Luciferase reporter assay: The Atg16l1 3′UTR luciferase construct vector (1 μg) was co-transfected into RAW264.7 cells with miR-20a mimic (100 nM) or miR-scramble (100 nM). The results show
significant repression of the 3′UTR of Atg16l-WT by miR-20a mimics. The data are shown as the mean ± SD of three independent experiments performed in triplicates. **P b 0.01 compared to
normoxia. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
302
303
304
305
antagomiR and bafilomycin as a control for autophagic flux in 306 RAW264.7 preosteoclast cells prior to hypoxia-induced osteoclast dif- 307 ferentiation (Fig. 3A).
Consistent with the reporter analysis of 3′UTR of 308 Atg16l1, overexpression of
To examine the functional role of the miR-20a in osteoclast differentiation, we miR-20a significantly reduced lipidation of 309
ectopically expressed the miR-20a by transfection mimics or
LC3 (LC3-II isoform) and ATG16L1 protein level by immunoblots 310
3.3. MiR-20a negatively regulates ATG16L1 of autophagy and osteoclast
differentiation
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
UNCORRECTED PROOF
8
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
311319
312
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
9
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
UNCORRECTED PROOF
10
α has six binding sites on mmu-miR-20a promoter region.
Fig. 4. HIF-1α mediates transcriptional silencing of the miR-20a during hypoxia-induced osteoclast differentiation. (A). HIF-1
(B). ChIP assay of RAW264.7 cells treated with shHIF-1α showed marked reduction in binding of HIF-1α with the miR-20a-5p promoter. The data are shown as the mean ± SD of
three independent experiments performed in triplicates. **
P b 0.01 compared to normoxia. (C). miR-20a was significantly upregulated in the presence of shHIF-1α under 24 h hypoxia,
as determined through miRNA Q-PCR. The data are shown as the mean ± SD of three independent experiments performed in triplicates.
P b* 0.05 compared to shLacZ under hypoxia.
α. (D). Inhibition of autophagy as demonstrated through reg
miRNA Q-PCR of osteoclast progenitor-RAW264.7 cells transduced with lentiviral vectors encoding control shLacZ or shHIF-1
ulation of early autophagy markers— Lc3, P62, Atg5 , Atg12, Atg16l1, Atg7, Becn1 and Atg9a (E). Downregulation of early osteoclast differentiation markers
—Traf6, Nfatc1, Ctsk, Cfos, Mmp9
and Trap. The data are shown as the mean ± SD of three independent experiments performed in triplicates.P *b 0.05 compared to shLacZ. (F). Cells were transduced with lentiviral vectors
α showed TRAP-negative cells. Data are mean ± SD, representative
encoding control shLacZ or shHIF-1α for 7 day osteoclast differentiation and analyzed for TRAPlevels. Cells with shHIF-1
of three independent experiments performed in triplicates. Magni
fication (100×); scale bar = 100μm. *P b 0.05 compared to shLacZ.
(Fig. 3B). Furthermore, the marker genes of osteoclast differentiation
and Atg16l1 decreased after transfection of miR-20a mimics (Figs. 3C,
D, E and F). Consistently, osteoclast formation demonstrated by TRAP
staining was inhibited by miR-20a mimics (Fig. 3 G). However,
antagomiR-20a would restore the inhibitory effect of miR-20a on osteoclast differentiation and formation (Figs. 3C, D, E, F and G). Our data
demonstrate that miR-20a reduces autophagy by targeting 3 ′UTR of
Atg16l1 and osteoclast differentiation.
313
3.4. HIF-1α suppresses miR-20a under hypoxia
Hypoxia regulates the expression of many genes via a highly con- 320
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
11
314
served transcription factor called hypoxia-inducible factor (HIF). To fur- 321
315
ther elucidate the role of HIF-1α on miRNA regulation, the binding sites 322
316
317
of HIF-1α located on miR-20a were determined using the Eukaryotic 323
Promoter Database (http://epd.vital-it.ch/) for upstream positions 324
318
of approximately −2000 to +100 bp. HIF-1α has six binding sites 325
Fig. 3. Inhibition of autophagy and osteoclast differentiation by miR-20a. (A) Relative expression of miR-20a in RAW264.7 cells transduced with miR-20a mimics or antagomiR under hypoxia (24 h).
The data are shown as the mean ± SD of three independent experiments performed in triplicates. (B) miR-20a downregulates expression of ATG16L1 and, LC3 under hypoxic environment (24 h) as
determined by western blot. The relative protein expressions were determined against β-actin. Relative expression of the osteoclast differentiation markers
Q3 (C) Nfatc1, (D) Traf6 and (E) Trap, and (F) autophagy marker Atg16l1 in osteoclast progenitor-RAW264.7 cells transduced with control scramble, miR-20a mimics or antagomiR. The data are shown as
the mean ± SD of three independent experiments performed in triplicates. *P b 0.05 compared to scramble. (G) RAW264.7 cells were transduced with control scramble, miR-20a mimics or antagomiR
as indicated and allowed to differentiate in the presence of RANKL and M-CSF. After 7 days, cells wereanalyzed for TRAP staining; cells with miR-20a mimics show TRAP-negative cells.
Magnification-100×; scale bar = 100 μm. Data are mean ± SD, representative of three independent experiments performed in triplicates. *P b 0.05 compared to miR-20a mimics.
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
on the promoter of miR-20a (Fig. 4A). The down-regulated miR-20a in
hypoxia could be a direct interaction between HIF-1α and miR20a.
degradation of long-lived proteins are severely impaired in ATG16L1- 390
deficient cells [29]. Since, ATG16L1 is crucial for autophagy process, 391 we in
the present study attempted to analyze miRNA target under hyp- 392 oxic
To determine whether miR-20a is directly regulated by HIF-1α, we
conditions through in-silico gene prediction method. We observed 393 involvement
performed loss-of-function experiments by lentiviral infection, viral vector
of 7 candidate miRNAs, of which miR-20a-5p showed sig- 394 nificant
containing short hairpin sequence knockdown to HIF-1α (shHIF-1α) or LacZ
suppression under hypoxic conditions. We further confirmed 395 that miR-20a
(shLacZ) as control to RAW264.7 cells. In addition, we confirmed
directly targets 3′UTR of Atg16l1 and repressed autophagy 396 by downregulating
transcriptional level binding by ChIP assay with HIF-1α antibody on miR-20a
autophagy-related proteins LC3 and ATG16L1. We 397 next focused on the effect
promoter. The binding of HIF-1α was markedly reduced at the miR-20a
of miR-20a on the expression levels of pro- 398 teins involved in osteoclast
promoter in the shHIF-1α cells (Fig. 4B) compared to shLacZ cells.
differentiation. MiR-20a under hypoxic con- 399 ditions significantly
Consistently, cellular expression of miR-20a was increased in shHIF-1α cells
downregulated the levels of Nfatc1, Traf6 and Trap 400 and reduced TRAP
by miRNA Q-PCR (Fig. 4C). We introduced
positive cells. Taken together, modulation of autoph- 401
shRNA via lentiviral transduction into macrophage cells RAW264.7,
agy and autophagy-related proteins by miRNA is a critical mechanism in 402
which are enriched for osteoclast progenitors, and then cultured them
hypoxia-induced osteoclast differentiation. 403
in RANKL/M-CSF-containing media to induce ex-vivo osteoclast differ- Studies show that lower O2 tension alters bone homeostatic mecha- 404
entiation. Functional analysis was then performed to evaluate 3 hallnisms, thereby leading to bone loss [30]. Hypoxia-induced osteoclast 405
marks of osteoclast differentiation: expression of the autophagy
bone resorption functions through highly conserved transcription factor 406
marker (Fig. 4D), expression of the osteoclast differentiation marker
(Fig. 4E) and TRAP staining for mature osteoclast (Fig. 4F). Knockdown
called hypoxia-inducible factor (HIF-α). During the normal oxygen ten- 407
sion, HIF-α is rapidly destroyed by prolyl hydroxylase domain enzymes 408
of HIF-1α significantly impaired autophagy-related proteins expression
and osteoclast differentiation, as evidenced by a mean 10% to 50% reduction in enucleation efficiency relative to scrambled shLacZ (Figs. 4D and
4E). Traf6 level and osteoclast differentiation decreased in shHIF-1α line
through proteasomal degradation. However, under hypoxic conditions, 409
HIF-1 α stabilizes and transactivates genes involved in adaptive re- 410
sponses [31,32]. Increased osteoclast differentiation and subsequent re- 411
sorption activity were observed under hypoxic conditions [33]. Studies 412
(Figs. 4E and F). These findings highlight that HIF-1α may suppress
demonstrate differential expression of hypoxia regulated miRNAs 413
miR-20a expression on transcriptional level during hypoxia-induced os(HRMs) under pathological conditions [34,35]. In the context of osteo- 414
teoclast differentiation. clast differentiation, Sugatani et al. [21] was first to identify the involve- 415 ment of microRNA (miR-223) in bone metabolism through
inhibition of 416
4. Discussion osteoclast differentiation in RAW264.7 cells. Recently, Lee et al. [8] have 417 showed that miR-124 suppresses NFATc1 expression and regulates os- 418
353 miRNA profiling of RAW 264.7 cells and monocyte/macrophage pre- teoclastogenesis in bone marrow macrophages. However, there is a 419
cursors shows hundreds of miRNA candidates involved in osteoclast diflack of clear understanding on the role of hypoxia–miRNA interaction 420
ferentiation [9,21]. However, the role of hypoxia in enhancing osteoclast
in osteoclast differentiation. We in the present study determined that 421
CORRECTED PROOF
326
327
differentiation through miRNA mediation remains elusive. Here, we
HIF-1α has a regulatory role by binding to the promoter of miRNA- 422
confirmed that the regulatory axis of HIF-1α–miRNA-20a–Atg16l1 is a 20a and represses its activity. Similar to our finding, He et al. showed 423
critical mechanism in hypoxia-induced osteoclast differentiation.
that HIF-1α down-regulates miR-17 and -20a by directly targeting 424
Hypoxia and inflammatory cytokines are involved in initiation and
p21 and STAT3 in myeloid leukemic cell differentiation [36]. We 425
progression of osteoclast differentiation. Interleukin-1 mediates TNF- showed experimental evidence that there was a significant reduction 426
induced osteoclastogenesis by enhancing the stromal cell expression in HIF-1α-miRNA-20a binding with concomitant increase in miR-20a 427
of RANKL [22], while hypoxia enhances osteoclastogenesis via increased
levels in the presence of shHIF-1α; thus shows clear evidence of HIF- 428
RANKL expression in the periodontal ligament of the compressed (hyp1α in negative regulation of miRNA-20a. From HIF-1α knockdown 429
oxic) side [23]. Growth differentiation factor 15, which is secreted from studies, we further confirmed effective down regulation of autopha- 430
osteocytes, significantly promotes osteoclast differentiation under
gy and osteoclast differentiation markers. Thus, these findings con- 431
hypoxic condition [24]. From the present study, we demonstrated that firm that HIF-1α could modulate miRNA including miR-20a in 432
hypoxia caused an increase in TRAP-positive cells with significant up- hypoxia-induced osteoclast differentiation. The interaction between 433
regulation in osteoclast differentiation markers. Further, we also identi- HIF-1α and miRNAs may be an important mechanism in osteoclast 434
fied that hypoxia caused significant enhancement in autophagy through differentiation. 435
regulation of autophagy-related proteins in RAW 246.7 cells in the
In summary, the present work highlights the role of miR-20a in 436
presence of RANKL and M-CSF. Previous study conducted by Zhao autophagy-related osteoclast differentiation under hypoxia stress. The 437
et al. reveals that autophagy regulates hypoxia-induced osteoclast dif- regulatory axis of HIF-1α–miR20a–Atg16l1 might be a critical mecha- 438
ferentiation through HIF-1α/BNIP3 signaling pathway in RAW264.7 nism for hypoxia-induced osteoclast differentiation. 439
cells [12]. Thus, induction of autophagy might be an important process Supplementary data to this article can be found online at http://dx. 440
in osteoclast differentiation under hypoxic condition.
doi.org/10.1016/j.bone.2014.11.026. 441
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026
12
376
377
K.-T. Sun et al. / Bone xxx (2014) xxx–xxx
The vital importance of miRNAs in the modulation of target genes involved in the autophagy pathway has recently been highlighted [25].
378
379
MiR-376b controls autophagy by directly targeting the intracellular
levels of two key autophagy proteins, ATG4C and beclin-1, in the 3′
380
UTR of their mRNAs [26]. MiR-101 suppresses autophagy by targeting rasrelated protein Rab-5A, stathmin 1, and ATG4D in MCF-7 breast cancer cells
to sensitize chemotherapy for cell death [27]. Furthermore, miR-21 regulates
osteoclastogenesis via down-regulating programmed cell death protein 4
(PDCD4) [21]. Gain and loss of miR-375 function reflect the regulation of
ATG7 expression by autophagy under hypoxic conditions [28]. In the present
study, we determined the candidate miRNA target for Atg16l1 gene, the
complex protein which regulates autophagy. ATG16L1 complex is involved in
LC3 lipidation for autophagosome formation. Both autophagosome formation
and
381
382
383
384
385
386
387
388
389
453
References
This study was supported by Research Laboratory of Pediatrics, 443 Children's
Hospital of China Medical University. This study was also 444Q9 supported by
China Medical University Hospital (Grant Numbers 445 DMR-102-019), National
Science Council (NSC-101-2314-B-039- 446 004-MYB), and Taiwan Department
of Health Clinical Trial and Re- 447 search Center for Excellence (Grant Numbers
MOHW103-TDU-B- 448
212-113002).
449Q10
450
Conflict of interest
None.
Acknowledgments
451
452
442
[19] Chien C-H, Sun Y-M, Chang W-C, Chiang-Hsieh P-Y, Lee T-Y, Tsai W-C, et al. Identi- 502 fying transcriptional start sites of human microRNAs based on high-throughput
se- 503
454
455
[1] Rodan GA, Martin TJ. Therapeutic approaches to bone diseases. Science 2000;289:
456
[2] Boyle WJ, Simonet WS, Lacey DL. Osteoclast differentiation and activation. Nature
457
2003;423:337–42. [21] Sugatani T, Vacher J, Hruska K. A microRNA expression signature of osteoclastogen- 507
458
[3] Hocking LJ, Whitehouse C, Helfrich MH. Autophagy: a new player in skeletal main- esis. Blood 2011;117:3648–57. 508 459 tenance? J Bone Miner Res 2012;27:1439–47. [22] Wei S, Kitaura H,
Zhou P, Ross FP, Teitelbaum SL. IL-1 mediates TNF-induced osteo- 509
460
461
462
463
464
[4] DeSelm C, Miller B, Zou W, Beatty W, van Meel E, Takahata Y, et al. Autophagy pro-
465
[6] Shibuya H, Nakasa T, Adachi N, Nagata Y, Ishikawa M, Deie M, et al. Overexpression tion of osteoclastic differentiation by growth differentiation factor 15 upregulated 515 466 of microRNA223 in rheumatoid arthritis synovium controls osteoclast differentia- in osteocytic cells under hypoxia. J Bone Miner Res 2012;27:938–49. 516
467
tion. Mod Rheumatol 2013;23:674–85.
468
469
[7] Chen C, Cheng P, Xie H, Zhou HD, Wu XP, Liao EY, et al. MiR-503 regulates osteoclas- 2018–25.
470
[8] Lee Y, Kim HJ, Park CK, Kim YG, Lee HJ, Kim JY, et al. MicroRNA-124 regulates oste- tion and mTOR inhibition-related autophagy by targeting ATG4C and BECN1. Au- 520 471 oclast
differentiation. Bone 2013;56:383–9. tophagy 2012;8:165–76. 521
quencing data. Nucleic Acids Res 2011;39:9345–56.
504
1508–14. [20] Lewis BP, Shih IH, Jones-Rhoades MW, Bartel DP, Burge CB. Prediction of mammalian 505
teins regulate the secretory component of osteoclastic bone resorption. Dev Cell
2011;21:966–74.
microRNA targets. Cell 2003;115:787–98.
506
clastogenesis. J Clin Invest 2005;115:282–90. 510
[23] Park HJ, Baek KH, Lee HL, Kwon A, Hwang HR, Qadir AS, et al. Hypoxia inducible 511
factor-1alpha directly induces the expression of receptor activator of nuclear 512
[5] Lian JB, Stein GS, van Wijnen AJ, Stein JL, Hassan MQ, Gaur T, et al. Micro RNA control
of bone formation and homeostasis. Nat Rev Endocrinol 2012;8:212–27.
factor-kappaB ligand in periodontal ligament fibroblasts. Mol Cells 2011;31:573–8. 513
[24] Hinoi E, Ochi H, Takarada T, Nakatani E, Iezaki T, Nakajima H, et al. Positive regula- 514
[25] Frankel LB, Lund AH. MicroRNA regulation of autophagy. Carcinogenesis 2012;33: 517
togenesis via targeting RANK. J Bone Miner Res 2014;29:338–47.
518
[26] Korkmaz G, le Sage C, Tekirdag KA, Agami R, Gozuacik D. miR-376b controls starva- 519
472 [9] Kagiya T, Nakamura S. Expression profiling of microRNAs in RAW264.7 cells treated [27] Frankel LB, Wen J, Lees M, Høyer-Hansen M, Farkas T, Krogh A, et al. microRNA-101 522 473 with a
combination of tumor necrosis factor alpha and RANKL during osteoclast dif- is a potent inhibitor of autophagy. EMBO J 2011;30:4628–41. 523
474
ferentiation. J Periodontal Res 2013;48:373–85.
[28] Chang Y, Yan W, He X, Zhang L, Li C, Huang H, et al. miR-375 inhibits autophagy and 524
475
476
477
[10] Dandajena TC, Ihnat MA, Disch B, Thorpe J, Currier GF. Hypoxia triggers a HIF-
478
[11] Utting JC, Flanagan AM, Brandao-Burch A, Orriss IR, Arnett TR. Hypoxia stimulates
479
480
osteoclast formation from human peripheral blood. Cell Biochem Funct 2010;28:
481
482
483
[12] Zhao Y, Chen G, Zhang W, Xu N, Zhu JY, Jia J, et al. Autophagy regulates hypoxia-
484
485
486
[13] Wu H, Wang F, Hu S, Yin C, Li X, Zhao S, et al. MiR-20a and miR-106b negatively reg-
488
489
490
491
130a targets ATG2B and DICER1 to inhibit autophagy and trigger killing of chronic [33] Knowles HJ, Athanasou NA. Acute hypoxia and osteoclast activity: a balance be- 538
493
494
495
[16] Ma L, Young J, Prabhala H, Pan E, Mestdagh P, Muth D, et al. miR-9, a MYC/MYCN- oxia with CSC and EMT and their relationship with deregulated expression of
496
[17] Tay Y, Zhang J, Thomson AM, Lim B, Rigoutsos I. MicroRNAs to Nanog, Oct4 and Sox2 17/20a directly targeting p21 and STAT3: a role in myeloid leukemic cell differenti497 coding regions
modulate embryonic stem cell differentiation. Nature 2008;455: ation. Cell Death Differ 2013;20:408–18.
498
499
1124–8.
[18] Chen PS, Su JL, Cha ST, Tarn WY, Wang MY, Hsu HC, et al. miR-107 promotes tumor 500 progression by targeting the let-7 microRNA in
mice and humans. J Clin Invest 2011;
121:3442–55.
mediated differentiation of peripheral blood mononuclear cells into osteoclasts.
reduces viability of hepatocellular carcinoma cells under hypoxic conditions. Gastro- 525
enterology 2012;143:177–87 [e8].
526
Orthod Craniofac Res 2012;15:1–9. [29] Saitoh T, Fujita N, Jang MH, Uematsu S, Yang BG, Satoh T, et al. Loss of the autophagy 527
protein Atg16L1 enhances endotoxin-induced IL-1beta production. Nature 2008; 528
456:264–8.
529
374–80. [30] Knowles HJ, Cleton-Jansen AM, Korsching E, Athanasou NA. Hypoxia-inducible factor 530
Cell Physiol 2012;227:639–48.
regulates osteoclast-mediated bone resorption: role of angiopoietin-like 4. FASEB J 531
2010;24:4648–59. 532
RECTED PROOF
induced osteoclastogenesis through the HIF-1alpha/BNIP3 signaling pathway. J
[31] Wang GL, Jiang BH, Semenza GL. Effect of protein kinase and phosphatase inhibitors 533
ulate autophagy induced by leucine deprivation via suppression of ULK1 expression
on expression of hypoxia-inducible factor 1. Biochem Biophys Res Commun 1995; 534
216:669–75.
535
in C2C12 myoblasts. Cell Signal 2012;24:2179–86. [32] Li H, Ko HP, Whitlock JP. Induction of phosphoglycerate kinase 1 gene expression by 536 487 [14] Kovaleva V, Mora R, Park YJ, Plass C,
Chiramel AI, Bartenschlager R, et al. miRNA- hypoxia. Roles of Arnt and HIF1alpha. J Biol Chem 1996;271:21262–7. 537
501
lymphocytic leukemia cells. Cancer Res 2012;72:1763–72.
tween enhanced resorption and increased apoptosis. J Pathol 2009;218:256–64.
539
[15] Chen JF, Mandel EM, Thomson JM, Wu Q, Callis TE, Hammond SM, et al. The role of [34] Crosby ME, Kulshreshtha R, Ivan M, Glazer PM. MicroRNA regulation of DNA repair 540
microRNA-1 and microRNA-133 in skeletal muscle proliferation and differentiation. gene expression in hypoxic stress. Cancer Res 2009;69:1221–9. 541 492 Nat Genet 2006;38:228–33. [35] Bao B,
Azmi AS, Ali S, Ahmad A, Li Y, Banerjee S, et al. The biological kinship of hyp- 542
activated microRNA, regulates E-cadherin and cancer metastasis. Nat Cell Biol
2010;12:247–56.
miRNAs and tumor aggressiveness. Biochim Biophys Acta 1826;2012:272–96.
[36] He M, Wang QY, Yin QQ, Tang J, Lu Y, Zhou CX, et al. HIF-1alpha downregulates miR-
Please cite this article as: Sun K-T, et al, MicroRNA-20a regulates autophagy related protein-ATG16L1 in hypoxia-induced osteoclast differentiation, Bone
(2014), http://dx.doi.org/10.1016/j.bone.2014.11.026